Blowing 2% out of the water

So if you were to apply a theoretical model of direct marketing that included behavioral algorithms with a predictive capability, is that sufficient to anticipate a potential buyers needs? To really nail this down, the behavioral parameters need to be anchored by supporting elements, such as demographic or financial data. Why these? The demographic data is relevant in terms of stage of life; people with children have very different purchase patterns from people right out of college, people who rent an apartment spend money differently that people who own a home, etc. The financial aspect is equally important, the person with a fat bank account views life differently than the person skating on thin ice, but more importantly, financial information is unambiguous, the numbers are very precise. Getting to those numbers is a different matter, but when you do get to them there’s little room for interpretation.

Combining all three of these into a predictive model can become an incredibly powerful marketing tool. You end up with “stage of life” (that is, demographic) information, capabilities (financial) information, and most importantly, why they do what they do, and therefore how to reach them (psychometric profiling). Taking this type of information and creating cluster-based profiles allows you to know who is coming to your website (assuming cookies are enabled), what they are interested in, in terms of stage of life, what their spend capacity is, and how specifically to connect with them. If you overlay ancillary analytics such as collaborative filtering (“people like you bought something like this”), this is as close to a no-brainer as marketing can get. The marketing message can become so accurate and personalized it becomes funny and scary at the same time. You can forget the 2% response rates that used to get direct marketers excited, now we’re talking about hit rates of over 70% for anyone doing on-line shopping.